Obesity is a risk factor for a number of adverse medical and psychosocial outcomes. Solving the obesity problem will require a much greater emphasis on prevention. Treatment research indicates that interventions focusing directly on the food environment (structured meals and control of food stimuli) have considerable promise for weight control. Prevention research has found little support for trying to modify characteristics (e.g., knowledge, beliefs, skills) of individuals, perhaps because this approach does little to modify the food environment. The proposed research would be implemented in a workplace cafeteria where employees eat year-round and where the nutritional composition of the food can be modified. The influence of two promising interventions on food intake, weight change, and related outcomes will be investigated. The first intervention consists of a training program aimed at reducing the energy density of participants' diets inside and outside of the cafeteria. The second intervention consists of financial incentives to encourage the consumption of cafeteria foods lower in energy density. Participants will be 195 Caucasian and African-American men and women between the ages of 21 and 65. They will have a BMI of 23-35 and will have characteristics associated with an increased risk of weight gain. Selection and intake of lunch foods will be measured with digital photography and cashiers' computerized records of food purchases. Cafeteria patrons will be randomly assigned to 1) a measurement-only condition; 2) an intervention designed to teach participants how to reduce the energy density of their diets; or 3) the reduced energy density intervention plus financial incentives for choosing cafeteria foods low in energy density. The potential moderating influence of individual differences characteristics on outcome will also be evaluated. Outcome measures will be administered repeatedly during the intervention and at 9- and 18-month follow-ups. The assessments will include measures of anthropomorphic and blood lipid variables, nutritional intake (both in and out of the cafeteria), physical activity, overeating, eating self-efficacy, and weight-related quality of life.